Article: 964 of rec.aviation.homebuilt
Path: newshost.ncd.com!ncd.com!olivea!spool.mu.edu!sol.ctr.columbia.edu!howland.reston.ans.net!gatech!rutgers!walter!heron!venky
From: venky@heron.bellcore.com (G A Venkatesh)
Newsgroups: rec.aviation.homebuilt
Subject: The Agony and the Ecstacy of kit-building - #3
Message-ID: <1993Jul9.174556.5358@walter.bellcore.com>
Date: 9 Jul 93 17:45:56 GMT
Sender: news@walter.bellcore.com
Organization: Bellcore
Lines: 158
Nntp-Posting-Host: heron.bellcore.com


I must thank all those folks that have sent me comments and suggestions or 
just appreciation for my earlier posts. They indeed help. I am sorry that I 
have not been able to reply by e-mail to each one of you, but hope that you 
will understand when I say that "I would rather be building".

venky

		    Getting your hands (and walls) dirty

				   or

			How to get your glass wet


  "Start with the bulkheads", the common wisdom says, but there aren't any 
bulkheads in the wing kit. Only flying surfaces. So where does one start? The 
suggested practice layups in the Composite Construction booklet from RAF 
seemed like a good idea. The first suggestion was to layup 6 layers of 
rectangular pieces of bi-directional cloth. The glass was cut with a level of 
concentration that in retrospect seems extremely ridiculous and epoxy poured 
and spread on each of the layers with more care than a chemist working with 
highly explosive ingredients. But it is the first layup and regardless of 
how many books you have read and how many videos you have watched, you don't 
know what the "envelope of correct procedure" is (in much the same way you 
are uncertain about what you can do with a plane when you know very little 
about its limits). You wonder about every strand that seems out of place; 
every stroke of the squeegee either seems too much or too little. Finally, 
you are done wetting it out and you realize that if you spent time 
proportional to what you did on that one little piece for the actual 
construction of the plane, the only type of medical certificate that you 
will have at the end is the one that says DECEASED.

The piece manages to cure despite trips every hour or two with random pokings
to see if it is still wet. Trimmed to a recommended size, it is to be weighed
and compared with the recommended weight. The allowable range is 10.5-12.5 oz.
with 11oz as the ideal. The piece weighs in at 11.3oz. Wow! Feelings of being a
natural genius at glassing linger for a while. The naivete of the inexperienced
is short-lived but facilitates wonderful bouts of euphoria.

The second practice layup is to cover a piece of foam with several layers of
bidirectional and unidirectional cloth, to gain some confidence in the 
strength of such a construction as the booklet says. It was at this time that
the newsletter from Velocity with a blurb on hard-shelling came in the mail. 
Not being aware of the envelope of standard practices, should one trust oneself
to try the new technique? The experiment of hard-shelling half the practice
foam core has been discussed in this newsgroup earlier so suffice it to say
that the conclusion was to pass on the hard-shelling. Just feeling a little
ambitious, I decided to shape the foam into an airfoil for the practice layup 
and that turned out be a good introduction to the intricacies of handling foam.

Getting used to working with foam takes a while. Sanding foam has to be done 
with deliberate, gentle strokes. Foam is extremely intolerant of sloppiness 
and will tear its gut out at the least bit of provocation. But like most 
material, you get the hang of it with time. You also learn to make and use a 
million different types of sanding blocks as appropriate for the job.

It is pure joy to hold in your hands, the first piece of sandwich construction
that you can imagine to be a piece of a plane. It is also rather humbling to 
discover at the same time, that in the process of glassing it, you goofed by 
putting the unidirectional fibers perpendicular to the leading and trailing 
edges rather than parallel to them as suggested in the booklet. Such a mistake
would be disastrous in the actual construction of a plane. Nevertheless, 
you can now convince the best of the skeptics about the strength of sandwich 
composite construction by knocking some sense into them...literally.

So finally the day arrives when you decide to start on the plane which 
coincidentally is also the day that you discover that the kit isn't a Heathkit.
Having decided to start on the winglets as the least critical and the simplest 
of the assemblies in the wing kit, the first step seems fairly simple - Place 
upper and lower cores with the outboard (flat) side on a flat surface and 
bond them together. No problem. When you put the two core pieces together, 
they mate perfectly at the join line BUT the lower piece is angling outboard 
about 15 degrees. If one were to actually glue it that way, the rudder that 
you cut out later will have a nice outboard bend in the bottom 4 or 5 inches. 
That doesn't seem right, or is it some newfangled design that they have come 
out with? Considering that there are no pictures in the manual to either 
confirm or deny the latter possibility, you go back and look at the videos
again, looking closely at every shot of a winglet. You wish that they would 
provide a scale model of the plane to consult for such profound questions. 
The winglet in the video doesn't seem to have such a tilt but then the winglet 
in the videos is for the standard Velocity not for the 173 which has some 
changes in its winglet design. So the project is abandoned until the call can 
be made to the factory for the first "support". So much for the starting day.

The answer for this dilemma turns out to be rather obvious (for the factory
folks, of course). Sand the mating surface on the bottom core to make it level
and then mate it. Of course, then the mating surfaces will have different 
surface dimensions and the upper core will slightly overlap the lower core 
at the mating line. But you can sand them and contour them to shape. And 
then you realize that you need to change your model of kit-building from 
something that was closer to assembling pre-molded plastic pieces as in model 
planes to something that is closer to building a plane with modeling clay. 

Having dealt with the first setback, it is time to fuss over every ding, crack
or *imperfection* that your mind can imagine on the core surface. You fill 
every such travesty with micro only to realize in retrospect that it would
have been much better to leave some of them alone rather than aggravate the 
problem with less than perfect sanding techniques later. It is at this time 
that you first begin to develop a keener sense for texture and surface finish 
in your daily life. Like noticing potholes on the road that could be filled 
with micro or the food tray in the cafeteria that will need some serious 
sanding to make it smooth. This intolerance for surface aberrations will 
only get worse as you progress into more sanding and filling.

Eventually, the core is ready for the glass and the first of the near-disasters.
No matter how much you think you are ready for the process, actually doing it
will never feel the way you imagined it will a million times in your head.
The problems in this case started with the mistaken assumption (which was 
formed after doing the practice layup) that you need to coat all the pores 
of the foam completely (and make them disappear) before laying the glass on. 
The result, of course, is too much slurry. It is only later that you remember 
the incongruous mention in one of the newsletters about some builders putting 
on too much slurry and the suggestion that it is stupid. After letting the 
slurry cure for a couple of hours, you put the skin on and spread the epoxy. 
While one is rather occupied spreading and squeegeing as much out as 
possible, the epoxy manages to land itself on most surfaces around the cores. 
But you never notice any of that, not even the tiny streams that are going 
over to the other side of the winglet core and increasing your surface 
preparation time significantly for skinning on that side. The slurry, being 
more in quantity than what it should be, manages to shift while squeegeing 
resulting in a surface that is somewhat uneven in places. It also results in 
the glass fibers not being wholly straight in some spots. But theren't any
air bubbles.

Things often look better after curing than what they look like before. But 
even that fact wasn't sufficient to suppress the strong feeling that I may
have destroyed my first masterpiece. Time to make a panic call to the
local EAA tech counselor and to load the winglet into the car and drive up
to his place the same day. He hmms and hrrmmphs and finally pronounces the
piece passable but barely. It will be safe but would be a bitch to finish
later because of the uneven surface. Consultations with the factory folks
later would confirm it too. These things are over-engineered for a reason.
There will always be builders like me. Regardless of how nice the completed
plane might look eventually, I will always remember that scraggly piece of
work that will remain underneath the cosmetic exterior.

That single skin must have provided some valuable lessons because skinning the
other surface of the winglet after burying the cable to the COM antenna
is done with surprising speed and with extremely satisfying results. Moreover,
making the second winglet not only turns out to be faster but *much* cleaner 
on the surroundings as well. And the results are nice enough for the 
aforementioned tech counselor to pronounce that my third winglet would be 
perfect. I like him though.

An important lesson learned:

It is entirely futile to prepare a schedule for doing things more than one 
day in advance. It is easy to get carried away allocating time and tasks - 
skin on Sat, sand on Sun morning, skin the other side on Sun evening, sand it 
on Mon evening, etc. Things just don't work that way. There is absolutely no 
way of accurately predicting how much time a step further ahead would take or 
even when you would be able to do it until you have completed the previous 
step. And of course, there is always the possibility of additional steps 
that were wholly unanticipated.

Next in the series: 
	Tackling the canard or why tech counselors make house calls.